We are working toward elucidating the role of retinoic acid receptors (RARs) in mammary development and carcinogenesis and understanding the role of genes that function upstream of RARs to regulate their activity. Using genetically engineered mouse models, we demonstrated that RARs function as rheostats that up or downregulate mammary cancer predisposition in an oncogene-dependent manner and obtained preliminary evidence that RARs can also function as on-off switches that initiate mammary carcinogenesis. Further, we demonstrated that the cellular retinol-binding proteinl (CRBP1) is a positive upstream regulator of RAR activity and that CRBP1/RAR promote cell differentiation and suppress cell growth by inhibiting the PI3K/Akt survival pathway through a novel mechanism. RARs also regulate mammary duct development during puberty, a key developmental phase regulated by genes such as EGFR, BRCA1, etc. Our continuation aim 1 will elucidate the cellular basis of this developmental effect (cell autonomous vs. paracrine vs. systemic); test the notion that retinoid (Am580) treatment will prevent/slow tumor development in MMTV-Wnt1 mice but accelerate tumor development in MMTV-Neu mice; and identify the mechanism through which RAR inhibits Wnt signaling. Aim 2 will test the hypothesis that gobal RAR inhibition is sufficient to initiate mammary carcinogenesis using transgenic models in which a dominant negative RAR is expressed in the mammary epithelium. Aim 3 will further define the mechanim whereby CRBP1, acting through RAR, inhibits PI3K activity and will test whether CRBP1 is a positive upstream regulator of RAR activity in vivo. The class of drugs referred to as retinoids are promising agents in cancer prevention and treatment but so far clinical trials have not identified specific cancer cohorts in which retinoid treatment improves overall survival. Our studies demonstrate that the effect of retinoic acid receptors (the target of retinoids) is dependent on the nature of the oncogenic pathway behind the disease and that in some cases retinoid treatment stands to be beneficial but in others it might be counterproductive. This has direct implications to future retinoid clinical trials in that it advocates that participating patients be carefully stratified based on the molecular characteristics of their cancer.